@gano I used to not understand what colored means to me, but I sure do now. I have had my speakers voiced everywhich way in my room mimicking frequency responses of all kinds. What I am using now is a far sight from neutral (it is overtly colored). I guess any speaker is colored in that it sounds different from any other and none is perfectly neutral. That said, I think of colored as meaning intentionally and substantially deviating from neutral (or perhaps substantially deviating from some sort of Harman curve). Saying something is colored immediately begs the question- how?
Well in my case I have boosted bass beliw 80hz +10db (as to not interfere with voices), a lower bass at 80hz 5.5db, a peak at 110hz 7.5db, a dip at 141hz to 6.5db, a jump mp to 9db at 150hz -340hz, a massive rise to 18db from 350-485hz, a slight rise to 18.5db from 509hz to 800hz, a relative fall to 9db from 824hz to 3khz, and fall to 3db from 3.09khz to 20khz.
This makes for a midcentric sound where vocals are extremely emphasized or forward, warm, chesty, and emphasized in a way that is way different than my default speaker voicing or neutral. Bass is strong but only below 80hz so it does not interfere with vocal clarity and emphasis. In terms of vocal tone and transients, my digital voicing is probably going to be much more like a classic tube system running some classic speakers (admittedly with some harmonic differences, but tonally).
This is colored in that it makes no pretense about being neutral or flat or accurate, it is intentionally voiced to sound full, rich, chesty, bloomy, emotional for vocals by comparison to neutral. Treble is present but much less emphasized than before. To me, this has the effect of making voices seem bigger, more present and more real, but also sacrifices some treble detail. Detail is still there, but compared to my default tuning I need to listen for small details rather than just getting them tossed out at me like my Kantas would do by default out of the box (which also might have been ‘colored’ in a different way). To me, my chosen colored presentation has the effect of making vocals on normal recordings of normal quality (and poor ones) sound tremendously more real (instead of too quiet, too restrained, or too thin), but comes at the expense of giving up some room air and ambient detail in all recordings, but generally this is never a negative except in the very best audiophile recordings (these still sound amazing, but less amazing than the could if I made different eq choices).
This is my sound and what I like. And no question it is not neutral or accurate. The only way I I found what i was looking for is through experimenting with a zillion eq profiles in dirac with the assistance of chat gpt. I doubt i would have ever arrived in this place through gear/speaker swaps, or if so, not for years. What I do not know.
From my perspective, I was very lucky in that Chat GPT and Dirac did more for me than 10 years of speaker auditions and gear swaps (if I would have been patient enough and well funded enough to be able to do them, which is unlikely).
I think that as long as one’s dragon doesn’t involve chasing absolute detail (then you are always going to need very revealing speakers and gear that needs to be auditioned) or dynamics (then obviously eq cannot replicate the efficiency required in speakers to produce them), one could likely accomplish a ton of realizing ‘your sound’ without auditioning any gear at all with nothing more than owning a capable amp with enough headroom for dirac eq, and a set of speakers with a good off axis frequency response that takes well to eq (and of course lots and lots of time to dial them in to taste, AI can essentially train you to be your own speaker designer, but only you can listen and then ask the questions necessary to make followup adjustments).
Now, most people are never going to do this, nor should they (they have lives), but for those geeked out enough and motivated enough to make all the gear swaps chasing a dragon, I think maybe they should consider it.
